User talk:Jitse Niesen/Archive1

Ah ... I didn't see your comment until today. Thanks for heading me in the right direction.

I will be sure to start the mcml article on the weekend, when I have time to coordinate all that I know concerning it in accordance to the Wikipedia rules.

As per original research, it has been done by many people, and I will make sure to include references.

Aznrocket

I'm just replying to your question on if my wikipedia entry is copyrighted. Andy Morton is deceased, so I'm not sure the status of the copyright. It's mentioned in many poker books and is pretty common knowledge in poker so I'm not sure how copyright applies.

Fekko

Nice job on the numerical ordinary differential equations page.

Thanks. Rednblu 19:09 19 Jul 2003 (UTC)

Hi there, and welcome to the 'pedia! Impressive work so far -- just one request: could you try to avoid hard line breaks within paragraphs? This makes text harder to edit, and breaks list formatting. --Eloquence 21:02 19 Jul 2003 (UTC)

Okay, I will try to remember that. Thanks for repairing it. Jitse Niesen 11:02 20 Jul 2003 (UTC)

Hmm, the ugly LaTeX thing is definitely unfortunate. I think using LaTeX in running text is the right thing to do though, and hopefully the ugliness will be fixed at some point. The math source is often nearly unreadable when marked up with <sup> tags and whatnot, not to mention that it ends up typesetting the variables in a standard italicized font rather than the math font, which looks odd. It also makes rendering to non-HTML formats, like a future print version, more ugly, since again the variables won't be properly rendered in the math font. --Delirium 23:33, Jun 23, 2004 (UTC)

Just to clarify, I'm not reverting back to it. By saying it's the right thing to do, I meant in theory, and hopefully also eventually in practice, when the layout problem gets fixed. --Delirium 00:24, Jun 24, 2004 (UTC)

---

Thanks for fixing Lebesgue integration. I did not notice I was edited and older version. MathMartin 00:06, 4 Sep 2004 (UTC)

I'm sure that what you discussed on my talk page should be appropriate :) Dysprosia 00:20, 10 Sep 2004 (UTC)

I replied to you kind comment re Simon Donaldson on my talk page (so I'm writing this in case you didn't set a watch on my talk; it Feel free to delete this comment when read!). I note your discusion above re in-line LaTeX. Billlion 21:29, 14 Sep 2004 (UTC)

numerical analysis

Hi, we have met on a few numerical analysis pages. If you have any spare time it would nice if you could check Chebyshev polynomials and especially Chebyshev polynomials#Polynomial interpolation for errors.MathMartin 20:55, 15 Sep 2004 (UTC)

I rewrote Chebyshev nodes but I am not really satisfied. Feel free to improve my notation. More generally I intend to rewrite most of the spline pages which are in a really bad shape and contribute to many article on numerical analysis in the next few weeks. As the material is new to me I will probably make some mistakes and often lack the necessary broader scope. It would be nice if you could keep a watchful eye on me.MathMartin 11:07, 19 Sep 2004 (UTC)

You seem to be writing faster than I can check! This is of course good as many numerical analysis pages are indeed far worse than I'd like. In my opinion, the main problem with Wikipedia is not that we don't have enough articles, but that they are not good enough, but that's another issue. I don't know that much about splines, but I'll do my best.

Yes did quite some editing today :).

Where did you get the ||·||₀ notation for maximum norms from? I'd use ||·||_∞ (see e.g. Lp space), but perhaps I'm too theoretically minded. Similarly, for me the space C₀[−1, 1] is the space of continuous functions f with f(−1) = f(1) = 0.

I have seen ||·||₀ in a book but now I believe it is a printing error as I could not find any use of this notation on the web. I will change it to ||·||_∞. As for C₀[−1, 1], I meant to write C⁰[−1, 1].

Hopefully, I'll soon find time to write a bit about the Lebesgue constant, which seems to be the thing you were leading up to. O yes, one last note: I personally think that references should be provided with every article, to help the reader and also out of honesty. You don't need to give a references for every statement (though I wouldn't mind if you did!).

I agree there should be more references. But unfortunetely at the moment I get my knowledge from the numerical math script I am studying so I am unable to provide references.

Your edits to Chebyshev nodes made the article much clearer. Dank je wel.MathMartin 22:20, 19 Sep 2004 (UTC)

Runge phemonena

Hi, I have a question concerning Runge's phemonena. When using Chebyshev nodes to interpolate a function we can minimize the interpolation error, but the interpolation error still increases when we increase the degree of the polynomial. Is this true ?MathMartin 19:58, 20 Sep 2004 (UTC)

Different matrix notation

Do you know of any difference between the matrix notation

{\begin{bmatrix}1&0\\0&5\end{bmatrix}}

and

{\begin{pmatrix}1&0\\0&5\end{pmatrix}}

Which one is more common ? MathMartin 17:28, 25 Sep 2004 (UTC)

Rate of convergence

Thanks for editing this page. I only created it as a quick hack because an anon added this text at a really inopportune place. I probably should have listed it on cleanup... anyway it looks nice now. Gadykozma 00:36, 20 Oct 2004 (UTC)

Hi Jitse. Did you check your reference for the definition of rate of convergence? I am quite sure you are right and my textbook is wrong, but I would like to see it.

Also, there are sequences which do not fit in that definition, for example, the sequence

x_{n}={\frac {1}{2^{n^{2}2^{n}}}}

does not fit the criterium for quadratic convergence, as one would get μ=0; but it does not converge with a faster rate either. What do you think? Oleg Alexandrov 04:42, 14 Jan 2005 (UTC)

The references all use mu > 0, except perhaps the Burden and Faires book (I don't feel like going to the library to check it). Note that the sequence 1 / 2^(n^2 2^n) is quadratic convergent under the "extended" definition even when we add the requirement mu > 0, because it is dominated by the quadratic convergent sequence 1 / 2^(2^n). -- Jitse Niesen 11:13, 18 Jan 2005 (UTC)

Burden and Faires does not use \mu >0. That book stinks! I just removed it from the references.

You are right about 1 / 2^(n^2 2^n), I was still thinking in terms of the restrictive definition. Oleg Alexandrov 22:26, 18 Jan 2005 (UTC)

Federigo Enriques

Oops. I had a bunch of windows open and must've pulled that by accident. It'll go back in a second. --DMG413 01:46, 20 Oct 2004 (UTC)

For the chage of bilinear interpolation.

Yup, sorry, I was confused by the diagram. The Interpolation formulas are right.

Regards,

GunRock 16:22, 25 Oct 2004 (UTC)

Changes in the statement of Weierstrass's theorem

I replaced real with complex because that gives, according to Walter Rudin's book Principles of Mathematical Analysis, the theorem as "originally discovered by Weierstrass". There are two reasons why I rely on that. The first one is that a large part of Rudin's mathematical research was about generalizing the (Stone-)Weierstass theorem, and the second one is that his book (the one mentioned above) is known to have hardly any errors (if at all; I've never heard of one) and typos. A polynomial over C has its coefficients in C.

The Bilinear Interpolation

Yes, now the graph is much better, but usually, I think a point with a name Q_m, should have a coordinate (x_m,y_m), that may make a better sense. Thanks very much for your earnest. :GunRock 03:24, 27 Oct 2004 (UTC)

The Bilinear Interpolation

Hi there, thanks, and the diagram of the interpolation should be changed either. Thanks! GunRock 06:30, Oct 28, 2004 (UTC)

Nifty. :) Dysprosia 12:37, 5 Nov 2004 (UTC)

Newton's method

I thought about it for a long while, and I think you are right, 'to find' is better than 'to finding'. --Oleg

Variable speed of light

I just looked at that article; it seems to be mostly patent nonsense. Generally I have decided to only make changes in an article if I'm prepared to spen time on rewriting entirely. If you make one change to an article and the rest of the article is nonsense, it makes me feel like I'm giving tacit assent to the rest. Of course that's not true, that's the nature of WP. But I'd rather stay away from it. Right now I have a huge backlog of things I would like to rewrite. At some point Quantum Measurement which is atrocious. And that statement at the beginning of Wavefunction collapse about Ashfar's experiment is also ridiculous. CSTAR 16:40, 14 Dec 2004 (UTC)

Newton-Cotes formulas

What you fixed was my dumb mistake. Sorry and thanks. I will pay more attention. Oleg Alexandrov 18:05, 12 Jan 2005 (UTC)

Essential spectrum

Discussion moved to Talk:Essential spectrum.

Thanks

As a newcomer to WikiPedia, I welcome your suggestions. Thank you for the suggestions -- as this was my first article, I will continue to add examples and proofs to articles already published. --Tygar

Trend Quotes

Should the second quote, made by the same person who made the third quote, be separately attributed to the quoter?

Nice job, thanks!! GT

Vandermonde

Oops. I didn't see the approach discussed on Vandermonde and a google search for 'vandermonde interpolation' (without any quotes) on wikipedia didn't turn anything up, so I figured I was safe :o Thanks for pointing it out, I'll just redirect it. CryptoDerk 16:02, Feb 3, 2005 (UTC)

Re:Quantum adiabatic algorithm

I responded in the quantum computer talk page. The whole thing is highly suspect.CSTAR 18:18, 10 Feb 2005 (UTC)

New Mathematics Wikiportal

I noticed you've done some work on Mathematics articles. I wanted to point out to you the new Mathematics Wikiportal- more specifically, to the Mathematics Collaboration of the Week page. I'm looking for any math-related stubs or non-existant articles that you would like to see on Wikipedia. Additionally, I wondered if you'd be willing to help out on some of the Collaboration of the Week pages.

I encourage you to vote on the current Collaboration of the Week, because I'm very interested in which articles you think need to be written or added to, and because I understand that I cannot do the enormous amount of work required on some of the Math stubs alone. I'm asking for your help, and also your critiques on the way the portal is set up.

Please direct all comments to my user-talk page, the Math Wikiportal talk page, or the Math Collaboration of the Week talk page. Thanks a lot for your support! ral315 02:54, Feb 11, 2005 (UTC)

Grammarbot is a nuisance

As you seem to have noticed, the grammarbot has gone berserk. I blocked it. I'm sure I'll get flack for it, so I'm soliciting support for my action beforehand. CSTAR 16:13, 6 Mar 2005 (UTC)

I got more flack than I expected. In a period of a few hours the bot had edited 15 pages on my watchlist, two of which I think were wrong and several questionable (changes a, ... , b to a, ..., b). The policy on bots I though was pretty clear: if there is any doubt about there behavior they should be blocked. But I'm not going top expend any more energy on this matter.CSTAR 16:13, 8 Mar 2005 (UTC)

Thanks for Conjugate gradient method

Today I ran into your reply on talk:iterative method. So, belately, thanks a lot for writing an article on the conjugate gradient. That article was indeed badly missing, and I did not feel myself qualified enough to write a good article on it in a reasonable amount of time. Thanks again! Oleg Alexandrov 02:04, 11 Mar 2005 (UTC)

Removal of Christopher Wren from List of mathematical topics (V-Z)

According to St Andrews, Wren has done some maths. I'm not sure it's enough to warrant his inclusion in the list of maths topics though. There is also a short discussion at Talk:List of mathematicians. Just for your information ... Jitse Niesen 12:51, 18 Mar 2005 (UTC)

Thanks. Put the guy back. One day some kind soul might write about his math contributions in his bio article. Oleg Alexandrov 17:02, 18 Mar 2005 (UTC)

Volleyball

This is just to thank you again for your comments on the Volleyball talk page. I had meant to drop a note here before in connection with a few corrections you made in some pages I posted on this topic. I really appreciate your interest. I believe there's a lotta work to be done in this area, because volleyball is still too poorly covered in Wikipedia. I sincerely hope you'll remain an active contributor in this field. vlad_mv 02:56, 20 Mar 2005 (UTC)

New Mathematics Project Participants List

Hi Jitse.

In case you didn't follow the discussion on Wikipedia talk:WikiProject Mathematics here: Wikipedia talk:WikiProject Mathematics#Reformat of Participants list, I'm writing to you to let you know that I've converted the "WikiProject Mathematics Participants List" into a table. It is now alphabetical, includes links to the participant's talk page and contribution list, and has a field for "Areas of Interest". I thought you might want to check and/or update your entry.

Regards, Paul August ☎ 16:36, Mar 22, 2005 (UTC)

P.S. Thanks for all your good work on the PlanetMath Exchange project ;-)

Sorry

I'm not quite clear rules, sorry about that. Could you please help me to delete it if it were irrelevant. Also Method of rank, I don't know how to delete it. Thank you so much Trieu 18:26, 30 Mar 2005 (UTC)

Interior point methods

Thank you for your comments. I'll try to expand what I have. What I find difficult is how to convey the ideas without being either too mathematical (and so unreadable) or too shallow (and so useless). It's probably a matter of practice...

About Nesterov-Nemirovskii: this is the spelling I find in Steve Wright's book (Primal-dual interior-point methods). I'll fix the misspelled istance in Interior point method. Unfortunately, I won't be able to help about this topic, as I don't know it at all.

Thanks Marcol.

Category: Mathematical Methods

Yes, I think I confused myself. In one sense, computing is a subset of mathematics, so any algorithm is nessesarily mathematical. But otoh: many algorithms to not involve things that we think of as mathematical - sorting and searching. There's considerable overlap in the respect that matematical methods are often implemented on computers.

Added to this is the point that Category: Mathematical Methods is pretty empty. It shouldn't be. Many math pages contain useful recipies for doing things, although ...

Hmm, ok. How about this:

by mathematical method, we mean "useful recipies for doing common tasks", usually involving performing a sequence of steps. This category groups together pages which not merely contain such recipies, but are specifically about them. Although mathematical methods are often computer implemented, they are not nessesarily so. Recipies relating primarily to computing (eg: searching and sorting) are in Category: Algorithms.

The relation between the two categories would be via link, rather than making one a subcategory of the other.

Pmurray bigpond.com 01:36, 19 Apr 2005 (UTC)

Pmurray bigpond.com 23:45, 19 Apr 2005 (UTC)

I'm currently of the opinion that Category: Mathematical methods should be deleted. Can you think of any article that should be in Category: Mathematical methods and not in Category: Algorithms?

Not really - I only touched it because I was browsing in the arrea. Bombs away!

Moscow and Rhind Mathematical Papyri

I refer you to User:Mark Dingemanse/Roylee and encourage you to revert all his and his anon. edits. I was going to do it, but it looked complicated. Wizzy…☎ 09:20, Apr 24, 2005 (UTC)

Simplex algorithm and Downhill simplex method

Thanks for linking these articles; I had never seen the latter. It does raise the question though what to do with the articles. I see two possibilities: either to have two separate pages dedicated to the simplex algorithm in LP and the Nelder-Mead/downhill simplex method, or to merge the articles. Prima facie, I prefer the first option as they are different algorithms. I'd be interested in your thoughts.

By the way, do you know how widespread the term downhill simplex method is? I only knew Nelder-Mead simplex method, but numerical optimization is not quite my field. -- Jitse Niesen 10:51, 28 Apr 2005 (UTC)

Unfortunately this is not my field at all! I just discovered the "downhill simplex method" article yesterday (it was called just "downhill simplex" at the time). What I "know" about it is just from a little Googleing ;-). But from what I was able to find out "downhill simplex method" seems to be (on the internet at least) the more common name for the "Nelder-Mead simplex method". And, for what it is worth, I would agree that Simplex algorithm and Downhill simplex method should be separate articles.

By the way I discovered "downhill simplex" while looking into Pearson distribution (they were created by the same editor), which demonstrates that one of the best ways to get a valid stub expanded is to list it on VFD ;-) Paul August ☎ 16:13, Apr 28, 2005 (UTC)

I believe you may have unwittingly helped a vandal. IP 62.254.0.38 seems to be an ISP proxy shared by many users, one of them being a persistent vandal mostly using stealth or submarine vandalism. You "fixed" his half-failed attempt at destroying the content of Downhill simplex method and redirecting it to Simplex algorithm even though it should, apparently, not. (I've reverted it.) ←^#6_{talk 00:40, 15 May 2005 (UTC)}[reply]

Hi

Hello Jitse. Thanks for writing on my talk page. I strongly agree with you and I apologize for my error. --ImpalerBugz 07:59, 4 May 2005 (UTC)[reply]

Votes for Deletion

Don't try to start a flamewar in Wikipedia.

MSTCrow 08:02, May 20, 2005 (UTC)

Limburgish and pitch accent

I noticed that you seem to be familliar with the Limburgish language. I've commented on its talk page on the article's questionable use of the term tonal language, mentioning that I've written a draft new version of the article pitch accent at User:Alarm/Pitch accent. Since this article mentions Limburgish as an example, I would really value any input from people familliar with it. If you have the time and energy to take a look, I would appreciate any comments and/or suggestions for expansion. / Alarm 13:51, 24 May 2005 (UTC)[reply]

I can find no mistakes in your discussion of the Limburgish language (I know little linguistics, though). However, I don't quite understand the sentence "In this definition, a language is pitch-accented if the position of an accented syllable or mora determines the tonal pattern of the whole word (the pitch of each syllable or mora, usually high vs. low) according to a set of rules." in the section Pitch-accented languages vs tonal languages. At least for Norwegian and Limburgish, it is not only the position of the accented syllable but also the quality of the accent that determines the tonal pattern of the whole word. Finally, I'd like to urge you to move the article into the main space. -- Jitse Niesen 09:12, 8 Jun 2005 (UTC)

Would you like to be an admin?

Hi Jitse. I see you are back. Hope you had a good trip. We missed your sharp eyes in here; but it seems you are back with a vengeance as just today you caught three bugs in pages I had either on the watchlist or actually edited. :)

In the meantime, Paul August got nominated for admin and got promoted. So, mathematicians' conspiracy to take over Wikipedia is advancing. :) Would you like to help with that?

That is, and now seriously, would you like to be an admin? I think you will make a great admin (I wish I had your cool head when it comes to dealing with other editors :) I think from the mathematicians's side there will be good support. So just let me know. Oleg Alexandrov 03:43, 8 Jun 2005 (UTC)

And of course you will have my support as well. Paul August ☎ 03:58, Jun 8, 2005 (UTC)

Give me a week to recover from my trip, do all the things that heaped up while I was away, and think of reasons why I should refuse. Paul, congratulations on your promotion and particularly on the unanimous vote; very impressive. Of course, I'd've voted for you if I hadn't been away. -- Jitse Niesen 08:55, 8 Jun 2005 (UTC)

Your edit of least squares was very misleading

I changed the paragraph you wrote to read as follows:

In regression analysis, one replaces the relation

f(x_{i})\approx y_{i}

by

f(x_{i})=y_{i}+\varepsilon _{i},

where the noise term ε is a random variable with mean zero. In linear regression, the function f sometimes has the form f(x) = ax + b, with a and b to be determined; the general case is called nonlinear regression. Warning: It is tempting, but grossly wrong, to think that the reason for the name linear regression is that the graph of the function f(x) = ax + b is a line. But in fact, fitting a curve f(x) = ax_{2 + bx + c, estimating a, b, and c by least squares, is an instance of linear regression as well, because the vector of least-square estimates of a, b, and c is a linear transformation of the vector whose components are f(x_i) + ε_i. See linear regression.}

No reputable statistician would endorse the view that I warned against here. Michael Hardy 15:20, 9 Jun 2005 (UTC)

Families

Hi, we have some discussion going on, all of them in some way connected to families. Maybe it's a cultural thing. My German professor defined a basis as a family and I found lots of German sources doing the same. But I actually found hardly any English sources doing this. Generally families don't seem too popular in the English math community. Mathworld uses the curly brackets notation for families, which I find misleading for the reasons stated in my article.

Families is a formalisation of statements as "let v₁, ..., v_n be a basis" or "[a matrix A is invertible, if] the columns of A are linearly independent" (taken from invertible matrix).

It is perfectly possible to do linear algebra without using index notation, I actually prefer index free notation

\int _{\Omega }f

over

\int _{x\in \Omega }f(x).

I like universal properties and there is no need for families when talking about a free module over a set, the corresponding basis.

But many people, especially physicists, prefer index notation. They like to talk about first, second, or third coordinates, they use matrices instead of homomorphisms. Index notation can be formulated exactly, but this requires using families.

Making index free definitions and having pages using index notation refering to those definitions leads to small inconsistencies as is the case with invertible matrix. Markus Schmaus 15:42, 16 Jun 2005 (UTC)

Stable polynomial

About the Wilkinson's polynomial: it is true that zeros are better conditionned when any polynomial (with distinct roots!) is expressed in its Lagrange form. I added this fact here because it seemed appropriate for 2 reasons:

this polynomial is usually defined by its lagrange form
its zeros are very ill-conditionned in the power basis form

But you might be right when saying it is not the best place to talk about the subject of conditionning of roots of polynomials (which doesn't seem to be treated anywhere...). Maybe we should create a new page on this topic (or a related topic).

My information comes from my own end-of-study work so we should find another source to cite (but I don't know any).

Thanks for your attention and corrections...Julien Tuerlinckx 13:04, 17 Jun 2005 (UTC)

-- I should have been more precise as the Lagrange basis is better... than the power basis. I will provide you with a proof soon. By the way, I'm thinking of writing an article about sensitivity of polynomials (for roots and evaluation) in a few days... But I don't think I'll include the proof in it as proofs are generally not convenient on wikipedia pages.

Anyway, I already made the Wilkinson's part more precise. I'm just wondering if we should not remove the f(z)=prod... part as it is now the same as above... Regards, Julien Tuerlinckx 15:54, 17 Jun 2005 (UTC)

OK I left a proof on my talk page, I hope you will be convinced. I can still provide more information if you ask me.

On Wilkinson:Sorry that my changes lead to a wrong article... Hopefully you corrected it. Julien Tuerlinckx 18:59, 17 Jun 2005 (UTC)

Hi, just noticed there was a page named Routh Hurwitz Stability Criterion. What should we do about that? Julien Tuerlinckx 12:21, 18 Jun 2005 (UTC)

In fact my third example was false as the roots of (z-1)^3 are all equal to 1 (and not -1 as I wrote it)! This polynomial is thus not stable. The anonymous user gave a correct example (I was confused with the necessary condition). He also changed the fourth example but the new and the old are both correct, so to me, no correction is necessary. Thanks for keeping me informed, cheers Julien Tuerlinckx 13:41, 22 Jun 2005 (UTC)

Addendum: In the case of a real quadratic polynomial positivity of coefficients is in fact sufficient for (Hurwitz-)stability (this fact could be included on the page?). A counterexample in degree 3 is p(x)=x^3+x^2+x+6=(x+2)(x^2-x+3) which has two complex roots with real part 1/2. --212.18.24.11 15:38, 22 Jun 2005 (UTC)

Open Access Bibliography

If you look at the Open Access Bibliography [1], you will find that it is relevant in every place that it has been added. It provides numerous references for every chosen topic, its content is freely available, and it is under a Creative Commons Attribution-Noncommercial License.

Symplectic integrator

Dear Jitse,

Thank you very much for your comment about a symplectic integrator on my talk page.

Actually, when I was writing that article, I had just a separable Hamiltonian in mind and didn't know such integrators as Störmer-Verlet method or the partitioned Lobatto IIIA/IIIB method. (I would very much like to get acquainted with these methods, so please let me know some references on these methods.)

I agree to define a symplectic integrator as any integrator that preserves the symplectic two-form, as you pointed out. I think that the article needs appropriate corrections and expansions. I wonder if you could kindly make these contributions.

I appreciate your kind comments again. I am eager to learn everything that I feel interesting including mathematics. Although there may be cultural differences among the fields of chemistry, physics, and mathematics, I hope that I could understand and obey a proper convention in each field.

Tot morgen!

NorioTakemoto 14:47, 21 Jun 2005 (UTC)